fix: update Long comparison table to verified single-model scores only
Browse files
README.md
CHANGED
|
@@ -121,9 +121,6 @@ MRE-T1 achieves state-of-the-art single-model performance on the [BRIGHT benchma
|
|
| 121 |
| **MRE-T1** | **~4B** | **35.1** |
|
| 122 |
| Google-Gecko-Text-Embedding-004 | — | 33.2 |
|
| 123 |
| gte-Qwen1.5-7B-instruct | 7B | 27.8 |
|
| 124 |
-
| SFR-Embedding-Mistral | 7B | 26.0 |
|
| 125 |
-
| GritLM-7B | 7B | 26.0 |
|
| 126 |
-
| e5-mistral-7b-instruct | 7B | 25.5 |
|
| 127 |
|
| 128 |
## Usage
|
| 129 |
|
|
|
|
| 121 |
| **MRE-T1** | **~4B** | **35.1** |
|
| 122 |
| Google-Gecko-Text-Embedding-004 | — | 33.2 |
|
| 123 |
| gte-Qwen1.5-7B-instruct | 7B | 27.8 |
|
|
|
|
|
|
|
|
|
|
| 124 |
|
| 125 |
## Usage
|
| 126 |
|